From owner-svn-src-all@FreeBSD.ORG Sun Jun 17 21:02:48 2012 Return-Path: Delivered-To: svn-src-all@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id DBBF9106564A; Sun, 17 Jun 2012 21:02:48 +0000 (UTC) (envelope-from phk@FreeBSD.org) Received: from svn.freebsd.org (svn.freebsd.org [IPv6:2001:4f8:fff6::2c]) by mx1.freebsd.org (Postfix) with ESMTP id C708E8FC0A; Sun, 17 Jun 2012 21:02:48 +0000 (UTC) Received: from svn.freebsd.org (localhost [127.0.0.1]) by svn.freebsd.org (8.14.4/8.14.4) with ESMTP id q5HL2mAW032401; Sun, 17 Jun 2012 21:02:48 GMT (envelope-from phk@svn.freebsd.org) Received: (from phk@localhost) by svn.freebsd.org (8.14.4/8.14.4/Submit) id q5HL2mG9032399; Sun, 17 Jun 2012 21:02:48 GMT (envelope-from phk@svn.freebsd.org) Message-Id: <201206172102.q5HL2mG9032399@svn.freebsd.org> From: Poul-Henning Kamp Date: Sun, 17 Jun 2012 21:02:48 +0000 (UTC) To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org X-SVN-Group: head MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Cc: Subject: svn commit: r237203 - head/sys/dev/fb X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 17 Jun 2012 21:02:49 -0000 Author: phk Date: Sun Jun 17 21:02:48 2012 New Revision: 237203 URL: http://svn.freebsd.org/changeset/base/237203 Log: On certain newer Intel Atom based motherboards, for instance the D2500CC which I have, syscons in text-mode fails to show the expected contents due to write errors into video-memory. At least one of the causes is that we copy from syscons internal buffer to the video memory with optimized bcopy(9) which uses >16bit operations. Until now, 32bit and wider operations have always worked on the video memory, but since I cannot find a single source which says that this SHALL work, and since these chipsets/bugs are now out there, this commit changes syscons to always use 16bit copies on i386 & amd64. This may be relevevant for PR's: 166262 166639 and various other bug reports floating elsewhere on the net, but I lack hardware to test those. Modified: head/sys/dev/fb/fbreg.h Modified: head/sys/dev/fb/fbreg.h ============================================================================== --- head/sys/dev/fb/fbreg.h Sun Jun 17 20:45:45 2012 (r237202) +++ head/sys/dev/fb/fbreg.h Sun Jun 17 21:02:48 2012 (r237203) @@ -35,9 +35,16 @@ /* some macros */ #if defined(__amd64__) || defined(__i386__) -#define bcopy_io(s, d, c) bcopy((void *)(s), (void *)(d), (c)) -#define bcopy_toio(s, d, c) bcopy((void *)(s), (void *)(d), (c)) -#define bcopy_fromio(s, d, c) bcopy((void *)(s), (void *)(d), (c)) + +static __inline void +copyw(uint16_t *src, uint16_t *dst, size_t size) +{ + while (size--) + *dst++ = *src++; +} +#define bcopy_io(s, d, c) copyw((void*)(s), (void*)(d), (c)) +#define bcopy_toio(s, d, c) copyw((void*)(s), (void*)(d), (c)) +#define bcopy_fromio(s, d, c) copyw((void*)(s), (void*)(d), (c)) #define bzero_io(d, c) bzero((void *)(d), (c)) #define fill_io(p, d, c) fill((p), (void *)(d), (c)) #define fillw_io(p, d, c) fillw((p), (void *)(d), (c))